Cargando…
An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network
Recently, deep learning approaches, especially convolutional neural networks (CNNs), have attracted extensive attention in iris recognition. Though CNN-based approaches realize automatic feature extraction and achieve outstanding performance, they usually require more training samples and higher com...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8197830/ https://www.ncbi.nlm.nih.gov/pubmed/34071850 http://dx.doi.org/10.3390/s21113721 |
_version_ | 1783706995425017856 |
---|---|
author | Liu, Guoyang Zhou, Weidong Tian, Lan Liu, Wei Liu, Yingjian Xu, Hanwen |
author_facet | Liu, Guoyang Zhou, Weidong Tian, Lan Liu, Wei Liu, Yingjian Xu, Hanwen |
author_sort | Liu, Guoyang |
collection | PubMed |
description | Recently, deep learning approaches, especially convolutional neural networks (CNNs), have attracted extensive attention in iris recognition. Though CNN-based approaches realize automatic feature extraction and achieve outstanding performance, they usually require more training samples and higher computational complexity than the classic methods. This work focuses on training a novel condensed 2-channel (2-ch) CNN with few training samples for efficient and accurate iris identification and verification. A multi-branch CNN with three well-designed online augmentation schemes and radial attention layers is first proposed as a high-performance basic iris classifier. Then, both branch pruning and channel pruning are achieved by analyzing the weight distribution of the model. Finally, fast finetuning is optionally applied, which can significantly improve the performance of the pruned CNN while alleviating the computational burden. In addition, we further investigate the encoding ability of 2-ch CNN and propose an efficient iris recognition scheme suitable for large database application scenarios. Moreover, the gradient-based analysis results indicate that the proposed algorithm is robust to various image contaminations. We comprehensively evaluated our algorithm on three publicly available iris databases for which the results proved satisfactory for real-time iris recognition. |
format | Online Article Text |
id | pubmed-8197830 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-81978302021-06-14 An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network Liu, Guoyang Zhou, Weidong Tian, Lan Liu, Wei Liu, Yingjian Xu, Hanwen Sensors (Basel) Article Recently, deep learning approaches, especially convolutional neural networks (CNNs), have attracted extensive attention in iris recognition. Though CNN-based approaches realize automatic feature extraction and achieve outstanding performance, they usually require more training samples and higher computational complexity than the classic methods. This work focuses on training a novel condensed 2-channel (2-ch) CNN with few training samples for efficient and accurate iris identification and verification. A multi-branch CNN with three well-designed online augmentation schemes and radial attention layers is first proposed as a high-performance basic iris classifier. Then, both branch pruning and channel pruning are achieved by analyzing the weight distribution of the model. Finally, fast finetuning is optionally applied, which can significantly improve the performance of the pruned CNN while alleviating the computational burden. In addition, we further investigate the encoding ability of 2-ch CNN and propose an efficient iris recognition scheme suitable for large database application scenarios. Moreover, the gradient-based analysis results indicate that the proposed algorithm is robust to various image contaminations. We comprehensively evaluated our algorithm on three publicly available iris databases for which the results proved satisfactory for real-time iris recognition. MDPI 2021-05-27 /pmc/articles/PMC8197830/ /pubmed/34071850 http://dx.doi.org/10.3390/s21113721 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Liu, Guoyang Zhou, Weidong Tian, Lan Liu, Wei Liu, Yingjian Xu, Hanwen An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title | An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title_full | An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title_fullStr | An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title_full_unstemmed | An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title_short | An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network |
title_sort | efficient and accurate iris recognition algorithm based on a novel condensed 2-ch deep convolutional neural network |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8197830/ https://www.ncbi.nlm.nih.gov/pubmed/34071850 http://dx.doi.org/10.3390/s21113721 |
work_keys_str_mv | AT liuguoyang anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT zhouweidong anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT tianlan anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT liuwei anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT liuyingjian anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT xuhanwen anefficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT liuguoyang efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT zhouweidong efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT tianlan efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT liuwei efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT liuyingjian efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork AT xuhanwen efficientandaccurateirisrecognitionalgorithmbasedonanovelcondensed2chdeepconvolutionalneuralnetwork |